Abstract High dose chemotherapy and hematopoietic stem cell transplant (HSCT) regimens offer prospects for cure in children and young adults with high risk and relapsed cancers. However, oncology and HSCT patients are at increased risk of developing treatment-related toxicity such as infections, oral complications, and deconditioning. Recent studies indicate the risk for BSIs, oral complications, and deconditioning can be reduced by participation in key activities of daily living (ADL), including daily bathing with chlorhexidine gluconate (CHG), physical activity and oral hygiene. Despite this, many patients fail to perform these important ADLs during their inpatient stay, at a time they are most susceptible. Our long-term goals are to identify key strategies to reduce treatment-related toxicity in oncology/HSCT patients, and spread these processes to other institutions and populations, to improve outcomes and survival. To promote adherence to these three key ADL, we developed the ?ADL 1-2-3? initiative which includes: daily CHG bath, at least two defined physical activities per day and oral care three times daily. We then pilot tested a token economy system to increase engagement in a small group of inpatient pediatric oncology/HSCT patients who were non-adherent with these target ADL 1-2- 3, which showed dramatic improvement in adherence. To further our efforts, we designed and built digital devices to support an electronic token economy system promoting ADL 1-2-3 adherence based on our pilot study. The proposed study will directly compare standard mechanisms to encourage ADL adherence in hospitalized oncology and HSCT patients to patient engagement through operant conditioning utilizing an automated ADL adherence system through a 1:1 randomized controlled study. Our central hypothesis is that patients using the automated token economy system will demonstrate higher ADL 1-2-3 adherence than patients receiving our current standard of care, and will be associated with lower BSI rates and improved clinical and patient reported outcomes. We will test this hypothesis through the following specific aims: Specific Aim 1: Compare the effectiveness of an automated ADL adherence system to standard of care at improving ADL 1-2-3 adherence through a randomized control trial. Specific Aim 2: Determine the impact of the automated token economy system on clinical outcomes and inpatient burden of care. We anticipate this work will identify effective strategies to increase adherence to key ADL in pediatric patients receiving intensive inpatient therapy, and reduce adverse outcomes including BSIs, mucositis, healthcare resource utilization and other treatment complications. We will demonstrate improved patient engagement resulting in greater self- management, and use these data to inform future design of large-scale trials. Finally, our data will likely be generalizable to other patient populations and settings of care.